OpenAI has launched API inference residency in India. Businesses and developers can now process supported GPT-6 Luna workloads inside the country.
The move targets data localisation rules. This matters most in regulated sectors such as financial services, where the location of data processing can decide whether firms deploy AI applications at all.
What the offering covers
Eligible customers can process API requests and run model computations in India. Supported customer content is also stored in the country.
The offering currently supports GPT-6 Luna input and output through OpenAI’s Responses and Chat Completions APIs. It covers direct API use. Organisations can use it for text-based work such as summarising documents, drafting replies and analysing information.
Customers switch it on by creating a new API project and choosing India as the region for both storage and processing. They must also use the matching India regional endpoint.
Why it matters to Indian firms
Nitin Bawankule, Head of Enterprise for India at OpenAI, said some Indian businesses have held back from using AI for important work. They need the data involved to be processed in the country. Local inference gives them more control over where supported workloads run, he said. It removes a practical barrier to adoption. It also gives Indian developers and startups more room to build for their own needs and for enterprises.
Building on data residency
The launch extends OpenAI’s existing data residency options in India. Those covered only where customer data is stored. With inference residency, both API request processing and model computation now happen in the country for supported workloads.
Part of a wider India push
The move builds on OpenAI’s wider India strategy. Earlier this year, the company introduced OpenAI for India. The initiative covers AI infrastructure, enterprise adoption, skills development and local partnerships. OpenAI also announced a partnership with the Tata Group to build domestic AI data centre capacity.
OpenAI recently brought a similar residency capability for certain models to Amazon Web Services Bedrock platform. It has also extended API inference residency to other markets. India, Australia and Japan are among the latest additions.
